There is a considerable interest in the development of photocatalytic CO2 conversion by sunlight since this process has similarities with natural photosynthesis on which life on Earth is based. At the moment, most of the efforts in this field have been aimed at increasing the productivity, rather than at the control of the product distribution. Particularly, compounds with two or more carbons (C2+) have higher added value than methane, carbon monoxide or formate that are typically the major products of CO2 reduction. This review focuses on those reports that have described the formation of compounds of two or more carbon atoms (C2+) in the photocatalytic CO2 reduction either by H2O or as H2 as source of electrons and protons. The existing literature has been organized according to the main factor considered as responsible for the selectivity to C2+ products, including photocatalyst structuration, co-catalyst nature, influence of defects, effects of surface plasmon band. Emphasis has been made on remarking the current 2 empirical knowledge based on experimental results and the lack of predictive capability that could lead to the development of efficient photocatalytic systems for C2+ production.

In order to increase the energy efficiency of dye-sensitized solar cells beyond 10%, an improved dye needs to be developed with greater light absorption in the red and near-infrared. Many dyes have been tested for this purpose; however, no dye with significant absorption beyond 750 nm has functioned properly. We have examined a series of ruthenium phthalocyanines, a dye class with large and tunable absorption in the red. For these dyes we observe a large reduction in the output voltage of the cells relative to the benchmark dye (N719). By examination of photovoltage transients and charge density measurements, we demonstrate that this reduction in voltage is caused by a 100-fold increase in the rate constant for recombination (iodine reduction) at the TiO2/electrolyte interface. N719, however, does not seem to catalyze this reaction. By examination of the literature, we propose that catalysis of the recombination reaction may be occurring for many other classes of potentially useful dyes including porphyrins, coumarins, perylenes, cyanines, merocyanines, and azulene. This widespread ability of the dye to catalyze recombination has not been appreciated before. This finding has important implications for future work to improve the red response of dye sensitized photovoltaics.